Precommitment Devices (Ulysses Contracts)
Binding your future self, and the research on what actually sticks
What are precommitment devices, and do they work?
A precommitment device is a constraint you impose on your future self while you still have the clarity to want the right thing — locking in a choice so the weaker, in-the-moment you cannot undo it. Commitment contracts (including ones with real money or social stakes) have solid experimental support, though they help most for people already motivated to change.
Willpower is unevenly distributed across time. The version of you setting a goal on Sunday is not the version of you facing the couch on Wednesday. Precommitment devices — named for Ulysses lashing himself to the mast so he could hear the Sirens without steering into them — exploit that gap on purpose: the strong-willed present binds the weak-willed future. Below are the core forms, the mechanism behind each, and an honest read on the evidence.
Practices
- Commitment contracts with real stakes
Put money or a consequence on the line so failing the goal costs something concrete.
- Anti-charity stakes
Pledge that failure sends your money to a cause you despise.
- Remove the option entirely
Delete the temptation from your environment so resisting it is never required.
- Self-imposed deadlines
Bind yourself to interim deadlines instead of one distant due date.
- Public commitment
Announce the goal to people whose opinion you care about.
- Cooling-off and waiting rules
Pre-commit to a mandatory delay before any impulsive action.
How these practices connect
Every one of these practices acts on the present self, while motivation is still high, to bind or box in a future self that present bias and depletable willpower will make weaker. None of them make the future self more disciplined in the moment of temptation; they either raise the cost of quitting (commitment-contracts, anti-charity-stakes, social-precommitment), remove the choice before it can be made (removing-the-option), pull a distant cost forward to where present-biased preferences still react (deadline-precommitment), or buy time for the urge itself to decay (cooling-off-rules). The common move is doing the work of self-control in advance, once, instead of repeatedly in the heat of the moment.
I keep breaking promises to myself, does telling other people about my goal actually help or just embarrass me if I quit?
Announcing a goal to people whose opinion you care about works through social-precommitment: quitting now would create a visible gap between what you said and what you did, and avoiding that gap becomes an external enforcer your private intentions didn't have. But there's a documented catch — some research suggests that announcing an identity-based goal like "I am writing a novel" can itself feel rewarding, giving you a taste of completion before you've done any of the work. The fix is framing the announcement as a binding promise you'll be checked on, not as an achievement to be congratulated for.
Is it better to put money on the line for a habit or just delete the app that's the actual problem?
These target different failure points. Commitment-contracts work through loss aversion — staking real money makes the prospect of losing it loom larger than the vague benefit of the habit, so you follow through to avoid the loss. Removing-the-option works on a different assumption entirely: that willpower is a depletable resource, so deleting the app removes the cue and the choice itself rather than betting on you winning the same fight against it every day. If the problem is a specific app or product you could genuinely live without, removing it is more robust than any stake, because it never requires you to resist anything at all.
Does giving my money to a charity I actually hate work better than just losing money if I fail my goal?
Anti-charity-stakes are built on top of commitment-contracts, not separate from them: a plain cash loss can be rationalized away as "it's only $20," but routing that same forfeit to a cause whose values you oppose adds an identity cost on top of the financial one, so failing now would mean actively funding something you stand against. That double sting is why anti-charity stakes are pitched as more motivating per dollar than a neutral stake. The honest caveat is that this specific variant has not been tested on its own in a randomized trial — it's a plausible amplifier of the studied commitment-contract effect, not an independently proven tactic.
Why do I always end up cramming a big project at the last minute even when I gave myself plenty of time?
This is present bias: a single distant deadline gets discounted so steeply in the moment that it functions as no deadline at all until the final days force a scramble, often at the cost of quality. Self-imposed deadlines address this by breaking the far-off due date into earlier, spaced, externally-known checkpoints, pulling the cost forward into a window where present-biased preferences still respond to it. The catch is that self-set deadlines, while better than none, reliably underperform ones enforced by someone else, because most people quietly build in too much slack when they're the only one holding themselves to it.
If I get a sudden urge to buy something impulsively, will just forcing myself to wait actually make the urge go away?
Yes, in a specific sense: the visceral pull of an impulse is time-limited and decays within minutes to hours, so a pre-set waiting rule like "24 hours before any non-essential purchase" is designed to let that hot reaction fade so the more deliberative part of your thinking can weigh in before the choice is locked. You're not out-willing the urge in the moment — you're outlasting it. What this doesn't do is address why the want showed up in the first place; it's a guardrail against acting on impulse, not a fix for the underlying urge itself.
Do these commitment tricks only work if I'm already motivated, or can they actually get me to start something I keep avoiding?
Commitment-contracts specifically lean on people who already want to change — take-up is self-selected, so they help someone follow through on a goal they've already decided matters, but they do comparatively little for someone who hasn't yet decided to act at all. Removing-the-option sidesteps this limitation somewhat, since it doesn't rely on willpower or motivation being present at the moment of temptation at all — once the app is deleted or the card is out of the house, the environment does the work regardless of how motivated you feel that day.
How do you choose between these?
Match the device to what's actually failing, not to what kind of person you think you are. If the temptation is a specific, removable thing — an app, a product, a card in your wallet — removing-the-option beats any stake or promise because it never requires you to resist anything. If the failure point is a distant, single due date, self-imposed deadlines address the specific timing problem by pulling the cost forward. If the failure point is sudden impulsive urges rather than a slow drift off a goal, cooling-off rules target that decay window directly. Save commitment-contracts and anti-charity-stakes for goals you've already decided to pursue but keep quietly letting slide, since their evidence is strongest for people who are self-selected into wanting the change already; add social-precommitment on top of any of these when you want an outside party enforcing the promise, but frame it as a binding commitment rather than an announcement to be praised for.
What these approaches do not settle
Anti-charity-stakes is the weakest-evidenced practice on this page: it has never been separately tested in a randomized trial, and everything said about it is an extrapolation from commitment-contract research plus a plausible mechanism, not its own proven result. There's also a real internal tension in social-precommitment — the same act of stating a goal publicly that is supposed to raise the cost of quitting can, per some research, instead deliver a premature sense of completion that lowers the odds you finish, which cuts directly against its own stated mechanism. And deadline-precommitment's own evidence caveat concedes that self-set deadlines underperform externally enforced ones, meaning the version of this practice a person can fully control themselves is also its least effective form.
Practice this with IX Coach
Articles that cover this
- Accountability Coaching, and the Statistic You Should Not Trust
- AI Coach vs. Accountability Partner: Which One Actually Gets You to Follow Through?
- Best Accountability App 2026: Habitica, StickK, Beeminder, Focusmate, and Coach.me Compared
- The Best Accountability App: How to Stay Accountable
- The Best Procrastination App: How to Choose One
- The Best App for Building Self-Discipline and Follow-Through
How this pairs with other practices
- Why a Time Log Alone Won't Change What You Do
Seeing the gap between your hours and your priorities is not the same as closing it. A binding precommitment is the second half.
- What to Do When a Plan Alone Won't Stop the Urge
An if-then plan works until the moment itself is loud enough to override it. Here's what to add when that keeps happening.
- Why Attention Management Needs a Ulysses Contract
Protecting focus in the moment fails the same way every time. A binding commitment made in advance is what actually holds when willpower runs out.
- Two Ways to Beat a Bad Moment Before It Happens
Environment Design nudges you toward the good option. A Ulysses Contract removes the bad one entirely. When each one actually holds.